gt, ge, lt, le, eq, ne, mat, nmat
.
The last two are for pattern matching. Logical operators: or
and
and
may be used as well as null
to indicate an empty value.
The value any
will match any field name. For numeric entries, you
may also use the comparison operators: >, >=, <, <=, ==, !=
(although,
I'm not certain that they are recommended and they usually require
extra delimiting). See the notes on delimiting special characters
in the section 6.2.
Following are some examples of RDB-type expressions of the type that might be used with BANAL's list commands, and their meaning. For more details read the perlre (1) man page.
If we have missed a special character then please let us know (you may still use the double delimiting method described below until a fix is made). You should realize, however, that you will still have to delimit the special characters on your command line (single delimit).
In the case that we have missed a special character use the following method to work around the problem.
When you find a special character (very hypothetical eg. '[') that you must delimit then you must delimit it once for your shell and you must delimit it again for perl (the perl set is the same as the Bourne shell). For example, if you want to match all invoices with a '[' in the Description field you must do:
bk invoices Description mat /\\\[/
or
bk invoices "Description mat /\[/"
You are using some books from a pre-0.03 release of BANAL. The quickest way to fix this is to edit the file, Company.rdb, in your books directory. Change the string 'Address1' to 'AddressA' and 'Address2' to 'AddressB'. These strings should be in the first line that doesn't begin with the character '#'. The name changes were made to assist in making a LaTeX driver for documents.
Remove the letter 'T' from the first line of bkd. This will also
disable some security checks. Please report this problem to:
Matthew.Rice@ftlsol.com
You must set your EDITOR environment variable. For example, to set your editor to /usr/local/bin/gnuclient, you would type:
export EDITOR=/usr/local/bin/gnuclient
To my line of thinking, a client is a person (you could be billing more than one person at the same company). I assume that anyone that you are billing is work. Thus that person should be contacted at their workplace. By keeping the company information seperate from the client it makes it easier to update the company information for all clients affiliated with said company. It also let's a client change companies without you losing the previous companies information after the client edit. Feel free to argue with me on this one.
You have to specify the directory for your books to the server. The server uses any rdb file that starts with a capital letter to generate edit and list functions for that database. You can specify the books' directory on the server's command line with the -b option or run the server in the books' directory.